Journal article
Integrating the sentiments of multiple news providers for stock market index movement prediction: A deep learning approach based on evidential reasoning rule
Information sciences, Vol.615, pp.529-556
11/2022
DOI: 10.1016/j.ins.2022.10.029
Abstract
In this study, we investigate the predictive capabilities of different news providers based on sentiment analysis, and propose a framework that endows different weights to different news providers for improving the prediction performance. In sentiment analysis, the prevalent Loughran-McDonald sentiment dictionary is utilized to calculate the sentiment scores of news articles, and the sentiment index of each news provider is obtained by integrating these sentiment scores. Based on the market data and sentiment indices of multiple news providers, we employ the recurrent neural network to build a number of base classifiers, and adopt the evidential reasoning rule to combine these base classifiers for predicting the stock market index movement. Additionally, the genetic algorithm is used to optimize the weights of base classifiers and important hyper-parameters of the recurrent neural network. In the experimental study, we apply the proposed approach to the daily movement prediction of the S&P 500 index, Dow Jones Industrial Average index and NASDAQ 100 index, and compare it with some state-of-the-art methods. The results show that our approach is effective for improving the prediction performance. Besides, the designed trading strategy based on the results of the proposed model achieves higher return rates than other trading strategies.
Details
- Title: Subtitle
- Integrating the sentiments of multiple news providers for stock market index movement prediction: A deep learning approach based on evidential reasoning rule
- Creators
- Ruize Gao - Chongqing UniversityShaoze Cui - Dalian University of TechnologyHongshan Xiao - Sichuan International Studies UniversityWeiguo Fan - Department of Business Analytics, Tippie College of Business, University of Iowa, Iowa City, IA 52242, United StatesHongwu Zhang - Chongqing UniversityYu Wang - Chongqing University
- Resource Type
- Journal article
- Publication Details
- Information sciences, Vol.615, pp.529-556
- Publisher
- Elsevier Inc
- DOI
- 10.1016/j.ins.2022.10.029
- ISSN
- 0020-0255
- eISSN
- 1872-6291
- Grant note
- DOI: 10.13039/501100001809, name: National Natural Science Foundation of China, award: 71471022, 71533001, 71801164; DOI: 10.13039/501100004543, name: China Scholarship Council, award: 202006060162; DOI: 10.13039/501100012166, name: National Key Research and Development Program of China, award: 2018YFB1403600; DOI: 10.13039/501100012226, name: Fundamental Research Funds for the Central Universities, award: 2021CDJSKJC10
- Language
- English
- Date published
- 11/2022
- Academic Unit
- Business Analytics
- Record Identifier
- 9984380521302771
Metrics
2 Record Views